Inspired by recent work in machine translation and object detection, we introduce an attention based model that automatically learns to describe the content of images. We describe how we can train this model in a deterministic manner using standard backpropagation techniques and stochastically by maximizing a variational lower bound. We also show through visualization how the model is able to automatically learn to fix its gaze on salient objects while generating the corresponding words in the output sequence. We validate the use of attention with state-of-the-art performance on three benchmark datasets: Flickr9k, Flickr30k and MS COCO.

/pdf/show-attend-and-tell-neural-image-caption-generation-with-2atzmg08ji.pdf

Show, Attend and Tell: Neural Image Caption Generation with Visual Attention

Inspired by recent work in machine translation and object detection, we introduce an attention based model that automatically learns to describe the content of images. We describe how we can train this model in a deterministic manner using standard backpropagation techniques and stochastically by maximizing a variational lower bound. We also show through visualization how the model is able to automatically learn to fix its gaze on salient objects while generating the corresponding words in the output sequence. We validate the use of attention with state-of-the-art performance on three benchmark datasets: Flickr8k, Flickr30k and MS COCO.

We present a model that generates natural language descriptions of images and their regions. Our approach leverages datasets of images and their sentence descriptions to learn about the inter-modal correspondences between language and visual data. Our alignment model is based on a novel combination of Convolutional Neural Networks over image regions, bidirectional Recurrent Neural Networks over sentences, and a structured objective that aligns the two modalities through a multimodal embedding. We then describe a Multimodal Recurrent Neural Network architecture that uses the inferred alignments to learn to generate novel descriptions of image regions. We demonstrate that our alignment model produces state of the art results in retrieval experiments on Flickr8K, Flickr30K and MSCOCO datasets. We then show that the generated descriptions significantly outperform retrieval baselines on both full images and on a new dataset of region-level annotations.

/pdf/deep-visual-semantic-alignments-for-generating-image-511hd6wg22.pdf

Deep visual-semantic alignments for generating image descriptions

Despite progress in perceptual tasks such as image classification, computers still perform poorly on cognitive tasks such as image description and question answering. Cognition is core to tasks that involve not just recognizing, but reasoning about our visual world. However, models used to tackle the rich content in images for cognitive tasks are still being trained using the same datasets designed for perceptual tasks. To achieve success at cognitive tasks, models need to understand the interactions and relationships between objects in an image. When asked "What vehicle is the person riding?", computers will need to identify the objects in an image as well as the relationships riding(man, carriage) and pulling(horse, carriage) to answer correctly that "the person is riding a horse-drawn carriage." In this paper, we present the Visual Genome dataset to enable the modeling of such relationships. We collect dense annotations of objects, attributes, and relationships within each image to learn these models. Specifically, our dataset contains over 108K images where each image has an average of $$35$$35 objects, $$26$$26 attributes, and $$21$$21 pairwise relationships between objects. We canonicalize the objects, attributes, relationships, and noun phrases in region descriptions and questions answer pairs to WordNet synsets. Together, these annotations represent the densest and largest dataset of image descriptions, objects, attributes, relationships, and question answer pairs.

/pdf/visual-genome-connecting-language-and-vision-using-15ih2ylk6r.pdf

Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations

Automatically describing an image with a sentence is a long-standing challenge in computer vision and natural language processing. Due to recent progress in object detection, attribute classification, action recognition, etc., there is renewed interest in this area. However, evaluating the quality of descriptions has proven to be challenging. We propose a novel paradigm for evaluating image descriptions that uses human consensus. This paradigm consists of three main parts: a new triplet-based method of collecting human annotations to measure consensus, a new automated metric that captures consensus, and two new datasets: PASCAL-50S and ABSTRACT-50S that contain 50 sentences describing each image. Our simple metric captures human judgment of consensus better than existing metrics across sentences generated by various sources. We also evaluate five state-of-the-art image description approaches using this new protocol and provide a benchmark for future comparisons. A version of CIDEr named CIDEr-D is available as a part of MS COCO evaluation server to enable systematic evaluation and benchmarking.

/pdf/cider-consensus-based-image-description-evaluation-2pnzkghk25.pdf

CIDEr: Consensus-based image description evaluation

This paper describes Meteor Universal, released for the 2014 ACL Workshop on Statistical Machine Translation. Meteor Universal brings language specific evaluation to previously unsupported target languages by (1) automatically extracting linguistic resources (paraphrase tables and function word lists) from the bitext used to train MT systems and (2) using a universal parameter set learned from pooling human judgments of translation quality from several language directions. Meteor Universal is shown to significantly outperform baseline BLEU on two new languages, Russian (WMT13) and Hindi (WMT14).

/pdf/meteor-universal-language-specific-translation-evaluation-1d9o0peq5a.pdf

Meteor Universal: Language Specific Translation Evaluation for Any Target Language

This paper describes Meteor 1.3, our submission to the 2011 EMNLP Workshop on Statistical Machine Translation automatic evaluation metric tasks. New metric features include improved text normalization, higher-precision paraphrase matching, and discrimination between content and function words. We include Ranking and Adequacy versions of the metric shown to have high correlation with human judgments of translation quality as well as a more balanced Tuning version shown to outperform BLEU in minimum error rate training for a phrase-based Urdu-English system.

/pdf/meteor-1-3-automatic-metric-for-reliable-optimization-and-lgnjkcwm1y.pdf

Meteor 1.3: Automatic Metric for Reliable Optimization and Evaluation of Machine Translation Systems

The Meteor Automatic Metric for Machine Translation evaluation, originally developed and released in 2004, was designed with the explicit goal of producing sentence-level scores which correlate well with human judgments of translation quality. Several key design decisions were incorporated into Meteor in support of this goal. In contrast with IBM's Bleu, which uses only precision-based features, Meteor uses and emphasizes recall in addition to precision, a property that has been confirmed by several metrics as being critical for high correlation with human judgments. Meteor also addresses the problem of reference translation variability by utilizing flexible word matching, allowing for morphological variants and synonyms to be taken into account as legitimate correspondences. Furthermore, the feature ingredients within Meteor are parameterized, allowing for the tuning of the metric's free parameters in search of values that result in optimal correlation with human judgments. Optimal parameters can be separately tuned for different types of human judgments and for different languages. We discuss the initial design of the Meteor metric, subsequent improvements, and performance in several independent evaluations in recent years.

/pdf/the-meteor-metric-for-automatic-evaluation-of-machine-1qpbic5v60.pdf

The Meteor metric for automatic evaluation of machine translation

We describe Sockeye (version 1.12), an open-source sequence-to-sequence toolkit for Neural Machine Translation (NMT). Sockeye is a production-ready framework for training and applying models as well as an experimental platform for researchers. Written in Python and built on MXNet, the toolkit offers scalable training and inference for the three most prominent encoder-decoder architectures: attentional recurrent neural networks, self-attentional transformers, and fully convolutional networks. Sockeye also supports a wide range of optimizers, normalization and regularization techniques, and inference improvements from current NMT literature. Users can easily run standard training recipes, explore different model settings, and incorporate new ideas. In this paper, we highlight Sockeye's features and benchmark it against other NMT toolkits on two language arcs from the 2017 Conference on Machine Translation (WMT): English-German and Latvian-English. We report competitive BLEU scores across all three architectures, including an overall best score for Sockeye's transformer implementation. To facilitate further comparison, we release all system outputs and training scripts used in our experiments. The Sockeye toolkit is free software released under the Apache 2.0 license.

Sockeye: A Toolkit for Neural Machine Translation.

Interest in neural machine translation has grown rapidly as its effectiveness has been demonstrated across language and data scenarios. New research regularly introduces architectural and algorithmic improvements that lead to significant gains over “vanilla” NMT implementations. However, these new techniques are rarely evaluated in the context of previously published techniques, specifically those that are widely used in state-of-the-art production and shared-task systems. As a result, it is often difficult to determine whether improvements from research will carry over to systems deployed for real-world use. In this work, we recommend three specific methods that are relatively easy to implement and result in much stronger experimental systems. Beyond reporting significantly higher BLEU scores, we conduct an in-depth analysis of where improvements originate and what inherent weaknesses of basic NMT models are being addressed. We then compare the relative gains afforded by several other techniques proposed in the literature when starting with vanilla systems versus our stronger baselines, showing that experimental conclusions may change depending on the baseline chosen. This indicates that choosing a strong baseline is crucial for reporting reliable experimental results.

/pdf/stronger-baselines-for-trustable-results-in-neural-machine-41egr25ftl.pdf

Michael Denkowski

Papers

Meteor Universal: Language Specific Translation Evaluation for Any Target Language

Meteor 1.3: Automatic Metric for Reliable Optimization and Evaluation of Machine Translation Systems

The Meteor metric for automatic evaluation of machine translation

Sockeye: A Toolkit for Neural Machine Translation.

Stronger Baselines for Trustable Results in Neural Machine Translation